HIV and T follicular helper cells: a dangerous relationship

Carola G. Vinuesa

J Clin Invest. 2012;122(9):3059-3062. https://doi.org/10.1172/JCI65175.

Commentary

HIV infection leads to progressive destruction of infected CD4 T cells, hypergammaglobulinemia, and loss of memory B cells. Germinal centers, which are key to memory formation and protective responses, are major HIV reservoirs in which the replicates within T follicular helper (TFH) cells. In this issue of the JCI, the Koup and Streeck groups report that chronic SIV/HIV infection promotes TFH cell accumulation, which may drive B cell dysregulation. Their discoveries suggest that HIV harnesses TFH cells to evade the antibody response.

Find the latest version: https://jci.me/65175/pdf commentaries HIV and T follicular helper cells: a dangerous relationship Carola G. Vinuesa

Department of and Immunity, John Curtin School of Medical Research, Australian National University, Canberra, Australia.

HIV infection leads to progressive destruction of infected CD4 T cells, HIV infects and expands TFH cells hypergammaglobulinemia, and loss of memory B cells. Germinal centers, It has been known since the late 1980s which are key to memory B cell formation and protective antibody respons- that germinal centers constitute the larg- es, are major HIV reservoirs in which the virus replicates within T follicular est reservoirs of HIV virions (9), which are helper (TFH) cells. In this issue of the JCI, the Koup and Streeck groups retained in the form of immune complexes report that chronic SIV/HIV infection promotes TFH cell accumulation, on the processes of follicular dendritic cells which may drive B cell dysregulation. Their discoveries suggest that HIV (FDCs) (10). It has also been suggested that harnesses TFH cells to evade the antibody response. active viral infection occurs primarily at these sites during clinical latency (11). Viral play an essential role in the pro- cells along two separate routes. An extra- replication does not occur in FDCs them- tection against viral infection, acting through follicular pathway generates short-lived, selves, but rather in germinal center CD4 different effector mechanisms to prevent viral low-affinity plasmablasts, whereas the T cells, now known as TFH cells (12–14). entry into host cells and eliminate infected follicular route gives rise to specialized TFH susceptibility to SIV infection is also cells. A few , however, have evolved microenvironments called germinal cen- confirmed in the current study by Petrovas sophisticated mechanisms to evade the neu- ters, in which affinity maturation of the and colleagues (5). There have even been tralizing antibody response and therefore antibody response takes place (7). Within suggestions that HIV-1 may replicate more persist. In the case of HIV-1, multiple mecha- germinal centers, -specific B cells actively in germinal center TFH cells com- nisms enable it to evade antibody binding acquire random point mutations targeted pared with that in other CD4 T cells (15). and replicate at leisure. These mechanisms at the Ig V region genes that change the In the context of murine LCMV infection, include the death of virus-specific CD4 T cells specificity of the B cell receptor. Those B viral persistence promotes the accumula- that are required to help the B cell response, cells that consequently bind antigen with tion of TFH cells in mouse secondary lym- a high mutation rate that allows the rapid higher affinity are thought to gain com- phoid tissues, which helps control infec- emergence of antigenic escape variants, and a petitive advantage to receive selection sig- tion (16). Now three groups also report structure of the envelope complex that makes nals from specialized TFH cells. Selected TFH cell accumulation during SIV and HIV conserved relatively inaccessible to B cells terminally differentiate into long- infection. Petrovas et al. observed that the antibodies (1, 2). In addition to preventing lived memory B cells or memory plasma proportion of TFH cells was increased by otherwise effective antibodies from inhibit- cells that home to the bone marrow (7). approximately 8 fold in 50% of chronically ing virus entry into cells, HIV can also alter Despite the described resistance of HIV SIV-1–infected rhesus macaques (RMs) — the quality of the antibody response: chronic to antibody-mediated neutralization, designated “TFHhi” — compared with that HIV infection is associated with hypergam- up to 25% of HIV-1–infected individuals in uninfected or acutely infected RMs or maglobulinemia, poor antibody responses to manage to develop high titers of broadly the “TFHlo” group of chronically infected protein vaccines, and decreased antigen-spe- neutralizing antibodies over time. The RMs (5). Hong and colleagues enumerated cific memory B cells (reviewed in ref. 3). The high-affinity human antibody VRC01, TFH cells per unit area of lymph node sec- studies in this issue of the JCI by Lindqvist which recognizes the initial site of CD4 tions; an approximately 4-fold increase in et al. (4) and Petrovas et al. (5), together with a attachment on HIV-1 GP120 and neutral- TFH cells was also observed in chronically parallel study by Hong et al. (6), suggest that izes about 90% of HIV-1 isolates, appears infected macaques compared with that in accumulation of T follicular helper (TFH) to evolve from a low-affinity unmutated uninfected or acutely infected ones (6). cells, together with possibly altered function germline ancestor via the acquisition of These results suggest an increase in TFH during chronic HIV infection, may be at the a vast number — 70 to 90 — of somatic cells in absolute numbers, as opposed to root of this form of humoral evasion. mutations (8). This represents around a a lower rate of depletion compared with 4-fold higher frequency of changes than other CD4 subsets. In the study by Germinal center origin of protective those typically observed during the affin- Lindqvist et al., HIV-infected individuals anti-HIV antibodies ity maturation process. These and other had approximately 10-fold higher frequen- B cells that receive cognate T cell help can data have suggested a germinal center cies of TFH cells compared with those of differentiate into antibody-producing origin and TFH-mediated selection of uninfected subjects (4). This is in itself broadly neutralizing antibodies and raise quite extraordinary since HIV kills the the possibility that altered selection in vast majority of cells it infects (17), and, Conflict of interest: The author has declared that no conflict of interest exists. germinal centers may contribute to the as discussed above, TFH cells themselves Citation for this article: J Clin Invest. 2012; failure to generate protective antibodies are susceptible to infection. None of the 122(9):3059–3062. doi:10.1172/JCI65175. in infected individuals. groups observed a correlation between

The Journal of Clinical Investigation http://www.jci.org Volume 122 Number 9 September 2012 3059 commentaries

Figure 1 Potential mechanisms by which TFH cells drive B cell dysregulation and constitute privileged virus reservoirs in HIV infection. (i) Excessive IL-6 produced by FDCs that harbor HIV viruses leads to upregulation of BCL-6 and increased differentiation and/or maintenance of TFH cells. Increased TFH cell numbers, together with abundant antigen, lowers competition for B cell selection reducing affinity maturation. (ii) Chronic virus infection changes the TFH cell expression of key molecules, including BCL6, IL6RA, IL4, LIGHT, and TGIF1, leading to altered function that may contribute to decreased memory B cell formation. (iii) In HIV-infected individuals, a larger fraction of TFH cells secrete IL-21 and possibly other known to promote B cell differentiation and antibody production; this can contribute to the hypergammaglobulinemia. (iv) The high IL-6 levels found in HIV infection may directly promote plasmacytosis and even enhance differentiation of memory B cells into plasma cells, leading to hypergammaglobulinemia and loss of memory B cells. (v) HIV replicates in TFH cells, and this may be enhanced by FDC-derived cytokines, including TNF. Exposure to high levels of IL-6 might also explain the increased proliferative rate of TFH cells in infected individuals, which may help maintain the viral reservoirs. (vi) It has been suggested that infection of TFH cells may be a strategy for viral persistence, since germinal centers are relatively devoid of cytolytic CD8+ T cells. GC, germinal center.

3060 The Journal of Clinical Investigation http://www.jci.org Volume 122 Number 9 September 2012 commentaries

HIV-1 plasma viral load and TFH cell num- with that in treated individuals (4). It is the TFH cell expansion seen in chronic bers, although it is possible that such cor- therefore possible that excess of this and HIV infection is beneficial or detrimental relation might exist with tissue virus levels; other TFH-derived cytokines, such as IFN-γ in mounting a high-affinity, broadly neu- virus is known to persist in tissues even and IL-10, contributes to hypergamma- tralizing antibody response against HIV in when it is undetectable in the blood (18). globulinemia (Figure 1, iii). Furthermore, humans. TFH cell expansion was shown to What drives the TFH cell accumulation the increased amounts of IL-6 found in help control chronic LCMV viral infection in chronic HIV infection? First, TFH cells chronic HIV infection are likely to increase in mice (16) and correlate with production extracted from HIV-infected individuals plasma cell survival and thus augment of high-affinity anti-SIV IgG antibodies in expressed more BCL-6 — the transcription total IgG levels (Figure 1, iv). macaques (5). However, finding that the factor required for TFH formation and Another intriguing observation was the TFH cell expansion in humans favored maintenance (19) — than those from treat- inverse correlation between the proportion accumulation of GAG-specific rather ed or uninfected subjects (4) (Figure 1, i). of TFH cells and memory B cells (4). There than envelope GP120-specific T cells (4) In the context of chronic LCMV viral infec- are at least two different explanations for — broadly neutralizing antibodies are typi- tion, TFH cell accumulation occurs as a this effect. First, it is likely that TFH cell cally directed against structures in enve- consequence of IL-6 produced by a radio- accumulation is accompanied by functional lope — suggests that this TFH cell response resistant cell (16), and FDCs are a plausible changes in TFH cells, and this may impair may not be terribly helpful in controlling candidate for this role within germinal cen- germinal center B cell selection and termi- the infection. The implications of the ters. IL-6 mediates upregulation of BCL6 on nal differentiation (Figure 1, ii). This is sug- recent discoveries for understanding HIV antigen-specific CD4 T cells (19). IL-6 has gested by the fact that the TFH/memory latency are also potentially exciting. Ger- been reported to be increased in HIV infec- B cell inverse correlation was only observed minal centers store large quantities of HIV tion (20), and levels are elevated in the plas- in HIV-infected individuals but not in unin- virus held on FDCs. The finding that TFH ma from infected macaques (5). Consistent fected individuals (4) and by the observed cells, which are susceptible to HIV infec- with a prominent role for IL-6 in driving differences in gene expression between tion, accumulate in lymph nodes rather TFH cell expansion, IL-6 receptor signal- TFH cells from infected and uninfected than being depleted, suggests these cells ing was increased in SIV-infected RMs with macaques (5). For example, the observed may constitute privileged latent reservoirs high frequency of TFH cells: IL-6Rα was upregulation of TGIF1, which limits (Figure 1, v). It has been suggested that the found expressed at increased levels on TFH TGF-β–responsive gene expression, may act paucity of SIV-specific CD8+ T cells in fol- cells (Figure 1, ii). Furthermore, TFH cells in a manner analogous to the depletion of licles may protect infected TFH cells from had increased phosphorylation of STAT3, follicular Tregs, which has been shown to being killed (ref. 6 and Figure 1, vi). Con- which mediates IL-6 receptor signaling (16). lead to expanded germinal centers contain- siderable effort is currently being directed ing increased numbers of non-antigen-spe- toward the development of treatments that Aberrant TFH expansion and cific GC B cells (22). Also, downregulation activate gene transcription to eliminate function may contribute to B cell of LIGHT, which costimulates CD40 trig- HIV latent reservoirs. It is conceivable that abnormalities gering (23), may dampen memory B cell for- these or other antiretroviral drugs could be Accumulation of TFH cells has been shown mation; CD40 signals have been reported to targeted to germinal center TFH cells by to maintain increased numbers of germi- bias the differentiation of germinal center coupling them to antibodies such as PD-1. nal center B cells and cause the prototypic B cells toward memory rather than plasma Finally, monitoring memory TFH-like cells systemic autoimmune disease SLE (21), cells (24). Increased expression of IL-6Ra in the blood has proven a useful biomarker probably due to the lowered threshold of B and BCL-6 in TFH cells is also likely to con- of TFH cells and germinal center dysregu- cell selection in germinal centers. Intrigu- fer functional changes. The other possibil- lation in secondary lymphoid tissues in the ingly, and as Moir and Fauci have pointed ity is that excessive IL-6 signals directly to context of autoimmunity (27). Given the out, virtually the same B cell abnormalities memory B cells and promotes their termi- parallels with chronic HIV infection, it is have been found in HIV and SLE (3). The nal differentiation into plasma cells (Figure likely that circulating TFH-like cells may explanation for this similarity may again 1, iv); precedents for this action of IL-6 have be a useful biomarker of the size of the HIV lie in the accumulation of TFH cells: the been previously described (25). Therefore, chronic reservoirs and provide a rationale three studies discussed here noted that the it is possible that memory B cell formation for using TFH-targeted therapies. increase in TFH cells in chronically infect- is intact, but further differentiation into ed subjects was accompanied by parallel plasma cells causes memory B cell depletion Acknowledgments increases in germinal center B cells. Also, a (Figure 1, iv). C.G. Vinuesa is supported by NHMRC positive correlation was observed between project and program grant funding and an total IgG and both the proportion of TFH Perspectives NHMRC Elizabeth Blackburn Fellowship. cells and BCL-6 expression on these cells Collectively, these studies suggest that over- (4). TFH cell accumulation during HIV activity and possibly dysregulation of TFH Address correspondence to: Carola G. infection, together with excess antigen cells explains the crippling B cell defects Vinuesa, Department of Pathogens and held on FDCs, may lower the threshold associated with HIV infection. It’s also Immunity, John Curtin School of Medical for B cell selection in germinal centers and tempting to speculate that this TFH cell Research, Australian National University, lead to the emergence of low-affinity vari- abnormality underpins the overrepresenta- Building 131, Garran Road, Acton, Can- ants (Figure 1, i). Lindqvist et al. observed tion of autoimmune conditions and auto- berra, ACT 0200, Australia. Phone: 61.26. that a larger fraction of TFH cells secreted antibody development in HIV infection 1254500; Fax: 61.26.1254595; E-mail: Carola. IL-21 in untreated individuals compared (26). A key outstanding issue is whether [email protected].

The Journal of Clinical Investigation http://www.jci.org Volume 122 Number 9 September 2012 3061 commentaries

1. Johnson RP. Mechanisms of protection against Am J Pathol. 1989;134(2):373–383. infection. Nature. 2005;434(7037):1093–1097. simian immunodeficiency virus infection. Vaccine. 10. Keele BF, et al. Characterization of the follicular 18. Cohen J. HIV/AIDS research. Tissue says blood is 2002;20(15):1985–1987. dendritic cell reservoir of human immunodefi- misleading, confusing HIV cure efforts. Science. 2. Kwong PD, et al. HIV-1 evades antibody–medi- ciency virus type 1. J Virol. 2008;82(11):5548–5561. 2011;334(6063):1614. ated neutralization through conformational 11. Heath SL, Tew JG, Tew JG, Szakal AK, Burton GF. Fol- 19. Crotty S. Follicular helper CD4 T cells (TFH). Annu masking of receptor–binding sites. Nature. 2002; licular dendritic cells and human immunodeficiency Rev Immunol. 2011;29:621–663. 420(6916):678–682. virus infectivity. Nature. 1995;377(6551):740–744. 20. Kuller LH, et al. Inflammatory and coagulation 3. Moir S, Fauci AS. B cells in HIV infection and dis- 12. Tenner-Racz K, von Stemm AM, Gühlk B, Schmitz biomarkers and mortality in patients with HIV ease. Nat Rev Immunol. 2009;9(4):235–45. J, Racz P. Are follicular dendritic cells, macro- infection. PLoS Med. 2008;5(10):e203. 4. Lindqvist M, et al. Expansion of HIV-specific T fol- phages and interdigitating cells of the lymphoid 21. Linterman M, et al. Follicular helper T cells are licular helper cells in chronic HIV infection. J Clin tissue productively infected by HIV? Res Virol. 1994; required for systemic autoimmunity. J Exp Med. Invest. 2012;122(9):3271–3280. 145(3–4):177–182. 2009;206(3):561–576. 5. Petrovas C, et al. CD4 T follicular helper cell 13. Haase AT. Population biology of HIV-1 infection: viral 22. Linterman MA, et al. Foxp3(+) follicular regulatory dynamics during SIV infection. J Clin Invest. 2012; and CD4+ T cell demographics and dynamics in lym- T cells control the germinal center response. Nat 122(9):3281–3294. phatic tissues. Annu Rev Immunol. 1999;17:625–656. Med. 2011;17(8):975–982. 6. Hong JJ, Amancha PK, Rogers K, Ansari AA, Vil- 14. Gratton S, Cheynier R, Dumaurier MJ, Oksen- 23. Duhen T, Pasero C, Mallet F, Barbarat B, Olive D, linger F. Spatial alterations between CD4(+) T hendler E, Wain-Hobson S. Highly restricted spread Costello RT. LIGHT costimulates CD40 triggering follicular helper, B, and CD8(+) T cells during of HIV-1 and multiply infected cells within splenic and induces immunoglobulin secretion; a novel key simian immunodeficiency virus infection: T/B cell germinal centers. Proc Natl Acad Sci U S A. 2000; partner in T cell–dependent B cell terminal differen- homeostasis, activation, and potential mechanism 97(26):14566–14571. tiation. Eur J Immunol. 2004;34(12):3534–3541. for viral escape. J Immunol. 2012;188(7):3247–3256. 15. Thacker TC, et al. Follicular dendritic cells and 24. Arpin C, et al. Generation of memory B cells and 7. Vinuesa CG, Tangye SG, Moser B, Mackay CR. Follic- human immunodeficiency virus type 1 transcrip- plasma cells in vitro. Science. 1995;268(5211):720–722. ular B helper T cells in antibody responses and auto- tion in CD4+ T cells. J Virol. 2009;83(1):150–158. 25. Beagley KW, et al. Peyer’s patch B cells with memo- immunity. Nat Rev Immunol. 2005;5(11):853–865. 16. Harker JA, Lewis GM, Mack L, Zuniga EI. Late inter- ry cell characteristics undergo terminal differentia- 8. Wu X, et al. Focused evolution of HIV-1 neutral- leukin-6 escalates T follicular helper cell responses tion within 24 hours in response to interleukin-6. izing antibodies revealed by structures and deep and controls a chronic viral infection. Science. . 1991;3(2):107–116. sequencing. Science. 2011;333(6049):1593–1602. 2011;334(6057):825–829. 26. Zandman-Goddard G, Shoenfeld Y. HIV and auto- 9. Ringler DJ, et al. Cellular localization of simian 17. Mattapallil JJ, Douek DC, Hill B, Nishimura Y, Martin immunity. Autoimmun Rev. 2002;1(6):329–337. immunodeficiency­ virus in lymphoid tissues. I. M, Roederer M. Massive infection and loss of memory 27. Vinuesa CG, Cook MC. Blood relatives of follicular Immuno­histochemistry and electron microscopy. CD4+ T cells in multiple tissues during acute SIV helper T cells. Immunity. 2011;34(1):10–12.

3062 The Journal of Clinical Investigation http://www.jci.org Volume 122 Number 9 September 2012